Gliomas account for almost half of all the primary brain tumors diagnosed in the U.S. Although these tumors are rare, survival rates are low (approximately 30% for 5-years). The only established risk factor for gliomas is ionizing radiation at therapeutic doses, but this factor can only account for a fraction of all brain tumors. Several dietary factors have been associated with the risk of gliomas, but findings are inconsistent, and most epidemiologic studies have relied on case-control designs. Unfortunately, case-control studies on brain tumors are particularly prone to biases because of the high fatality rates, and consequently, interpretation of findings from these studies is difficult. We propose to draw on three large prospective cohorts, the Health Professionals Follow-Up study, and the Nurses' Health Studies I and II to examine the relation between intakes of meat, fruits and vegetables and vitamins and the risk of gliomas. Detailed information on dietary intake was collected at baseline with food frequency questionnaires in each of the three cohorts and was updated periodically over the follow-up period. Data on detailed lifestyle factors, including smoking history, are available for the proposed analyses. We will use Cox proportional hazard models to estimate relative risks while adjusting for total energy intake and other potential confounders. This study will represent the first to examine dietary data obtained prior to brain tumor diagnosis. By combining three large cohorts, we will have approximately 321 incident cases of gliomas for the dietary analyses. Because the etiology of adult gliomas remains largely unknown, there are currently no modifiable risk factors which can be translated into prevention. This study should contribute substantially to our understanding of the relation between diet and the risk of adult gliomas, and might provide opportunities for prevention of this deadly disease.